Detection of unusual events in video streams, for the purpose of investigation and security is a challenging technology in crowded scenes. To address this issues, an algorithm is proposed, which is based on Histogram of Optical Flow Orientation image descriptor and nonlinear one-class SVM classification method. The optical flow method is computed at each pixel to extract the low-level features. Histogram of Optical Flow Orientation descriptor encoding the global moving information of each frame and one-class support vector machine classifier detects the unusual events in the current frame, after learning period distinguishing the common behaviors of the training frame. k nearest neighbor classifier is used to classify the abnormal frames in video streams. Further, by combining the background subtraction step and optical flow computation, a improved version of the detection algorithm is designed. This proposed method works on several benchmark datasets to detect unusual events. Histogram of optical flow orientation with nonlinear one-class SVM classifier shows the high performance result than, k nearest neighbor classifier with histogram of optical flow orientation.

Unusual event detection, Optical Flow, Histogram of Optical Flow Orientation (HOFO), one-class SVM, k nearest neighbor (kNN).

[1] Tiang Wang and Hichem Snoussi,“Detection of abnormal visual events via global optical flow orientation histogram orientation histogram”, IEEE Trans.vol. 9. No 6, june 2014.
[2] G. Lavee, E. Rivlin, and M. Rudzsky, “Understanding video events: A survey of methods for automatic interpretation of semantic occurrences in video”, Technion-Israel Inst. Technol., Haifa, Israel, Tech. Rep. CIS2009-06, 2009.
[3] G. Lavee, E. Rivlin, and M. Rudzsky, “‘Understanding video events: A survey of methods for automatic interpretation of semantic occurrences in video”, IEEE Trans. Syst. Man, Cybern. C, Appl. Rev., vol. 39, no. 5 , pp. 489504, Sep. 2009.
[4] T.Wang, H. Snoussi, and F. Smach, “Detection of visual abnormal events via one-class SVM”, in Proc. Int. Conf. Pattern Recognit. IPCV, vol. 1. 2012, pp. 113119.
[5] D. Kosmopoulos and S. P. Chatzis, “Robust visual behavior recognition”, IEEE Signal Process. Mag., vol. 27, no. 5, pp. 3445, Sep. 2010.
[6] Utasi and L. Czni, “Detection of unusual optical flow patterns by multilevel hidden Markov models”, Opt. Eng., vol. 49, no. 1, p. 017201 , 2010.
[7] T. Xiang and S. Gong, “Incremental and adaptive abnormal behaviour detection”, Comput. Vis. Image Understand., vol. 111, no. 1, pp. 5973 , 2008.
[8] T. Xiang and S. Gong, “Video behaviour profiling and abnormality detection without manual labelling”, in Proc. IEEE 10th ICCV, vol. 2. Oct. 2005, pp.12381245.
[9] T. S. Haines and T. Xiang, “Delta-dual hierarchical Dirichlet processes: A pragmatic abnormal behaviour detector”, in Proc. IEEE ICCV, Nov.2011, pp. 21982205.
[10] S. Kwak and H. Byun, “Detection of dominant flow and abnormal events in surveillance video”, Opt. Eng., vol. 50, no. 2, pp. 027202-1027202-8 , 2011.
[11] Y. Benezeth, P.-M. Jodoin, and V. Saligrama, “Abnormality detection using low-level co-occurring events”, Pattern Recognit. Lett., vol. 32, no. 3, pp. 423431, 2011.
[12] C. Bregler, “Learning and recognizing human dynamics in video sequences”, in Proc. IEEE Conf. CVPR, Jun. 1997, pp. 568574.
[13] T. Starner and A. Pentland, “Real-time American Sign Language recognition from video using hidden Markov models”, in Proc. Int. Symp. Comput. Vis., Coral Gables, FL, USA, 1995.
[14] A. F. Bobick and J. W. Davis, “The recognition of human movement using temporal templates”, IEEE Trans. Pattern Anal. Mach. Intell., vol. 23, no. 3, pp. 257267, Mar. 2001.
[15] Y. Chen, G. Liang, K. K. Lee, and Y. Xu, “Abnormal behavior detection by multi-SVM-based Bayesian network”, in Proc. ICIA, Jul. 2007, pp.298303.
[16] C. Schuldt, I. Laptev, and B. Caputo, “Recognizing human actions: A local SVM approach”, in Proc. 17th ICPR, vol. 3. 2004, pp.3236.
[17] Prof. Archana. V.Potnurwar and Dr. Mohammad Atique, “Visual Attention Key Frame Extraction for Video Annotations”, Int Journal of computer science and Engineering, vol.3, No.01, Page No (39-42), Jan 2014.
[18] PETS, Vellore, India. (2009). Performance Evaluation of Tracking and Surveillance (PETS) 2009 Benchmark Data. Multisensor Sequences Containing Different Crowd Activities[Online].Available:http://www.cvg.rdg.ac.uk/pets2009/a.html
[19] UMN, Minneapolis, MN, USA. (2006). Unusual Crowd Activity Dataset of University of Minnesota, Department of Computer Science and Engineering [Online]. Available: http://mha.cs.umn.edu/movies/crowdactivityall.avi
[20] B. K. Horn and B. G. Schunck, “Determining optical flow”, Artif. Intell.,vol. 17, no. 1, pp.185203, 1981.
[21] V. N. Vapnik and A. Lerner, “Pattern recognition using generalized portrait method”, Autom. Remote Control, vol. 24, no. 6, pp. 774780, 1963.
[22] B. E. Boser, I. M. Guyon, and V. N. Vapnik, “A training algorithm for optimal margin classifiers”, in Proc. ACM 5th Annu. Workshop COLT, Pittsburgh, PA, USA, Jul. 1992, pp. 144152.
[23] C. Piciarelli, C. Micheloni, and G. L. Foresti, “Trajectory-based anomalous event detection”, IEEE Trans. Circuits Syst. Video Technol., vol. 18 , no. 11, pp. 15441554, Nov. 2008.
[24] N. Cristianini and J. Shawe-Taylor, An Introduction to Support Vector Machines and Other Kernel-Based Learning Methods. Cambridge, U.K.: Cambridge Univ. Press, 2000.
[25] B. Schlkopf, J. C. Platt, J. Shawe-Taylor, A. J. Smola, and R. C. Williamson, “Estimating the support of a high-dimensional distribution”, Neural Comput., vol. 13, no. 7, pp. 14431471, 2001.
[26] Aswathy unnikrishnan, Ajesh F and Reshma S. Nair, “Detection of Abnormal Visual Events Using HOFO And KNN”, Int Journal of Informative and Futuristic Research, vol-02, Issue-09, Page No (1-15), May-2015.
[27] Chinnu A, “MRI Brain Tumor Classification Using SVM and Histogram Based Image Segmentation”, Int Journal of Computer Science and Information Technologies, vol.6(2), Page No (1505-1508), 2015.